Closed AndreaGarciaJuan closed 4 years ago
Don't forget to look at the pyXpcm issues ! This has been reported and fixed here: https://github.com/obidam/pyxpcm/issues/25
Fix on the master branch of pyXpcm, you should move away from stable pyXpcm version and work with the last dev version at master.
I am trying to save my dataset in a netCDF after doing a predict with option inplace=True:
m.predict(ds, features=features_in_ds, dim=features_zdim, inplace=True)
ds.to_netcdf('models/tests_ds_GLOBAL_REANALYSIS_PHY_001_030_temp.nc')
And I obteint this error:
TypeError: illegal data type for attribute b'_pyXpcm_cleanable', must be one of dict_keys(['S1', 'i1', 'u1', 'i2', 'u2', 'i4', 'u4', 'i8', 'u8', 'f4', 'f8']), got b1
Do you have an idea why this is happening?